In television receivers there is generally a sequence of separation of video signals. For example, an RF signal is detected, demodulated to provide a baseband signal. The baseband composite signal is again processed to provide luminance and chrominance signals. The chrominance signals are demodulated to provide color difference signals. After processing the baseband luminance signal and the color difference signals are converted to a final form for display, such as RGB.
In a picture-in-picture processor it is some times advantageous to provide inputs in composite form and to provide outputs in a mixed format, e.g., composite for the big picture and component for the inset or little picture. To combine these signals with a fast (multiplex) switch for further processing and display, it is necessary to first combine the components to composite form. The direct approach of simply summing the luminance and chrominance components, however, results in certain artifacts being produced in the inset picture.